Betalains and Phenolics in Red Beetroot (Beta vulgaris) Peel Extracts: Extraction and Characterisation Tytti Kujala*, Jyrki Loponen and Kalevi Pihlaja Department of Chemistry, Vatselankatu 2, FIN-20014 University of Turku, Finland. Fax: +3582-333 67 00. E-mail: [email protected] * Author for correspondence and reprint requests Z. Naturforsch. 56 c, 343-348 (2001); received January 9/February 12, 2001 Beta vulgaris , Betalains, Phenolics The extraction of red beetroot (Beta vulgaris ) peel betalains and phenolics was compared with two extraction methods and solvents. The content of total phenolics in the extracts was determined according to a modification of the Folin-Ciocalteu method and expressed as gallic acid equivalents (GAE). The profiles of extracts were analysed by high-performance liquid chromatography (HPLC). The compounds of beetroot peel extracted with 80% aqueous methanol were characterised from separated fractions using HPLC- diode array detection (HPLC-DAD) and HPLC- electrospray ionisation-mass spectrometry (HPLC-ESI-MS) tech­ niques. The extraction methods and the choice of solvent affected noticeably the content of individual compounds in the extract. The betalains found in beetroot peel extract were vulgaxanthin I, vulgaxanthin II, indicaxanthin, betanin, prebetanin, isobetanin and neobe- tanin. Also cyclodopa glucoside, /V-formylcyclodopa glucoside, glucoside of dihydroxyindol- carboxylic acid, betalamic acid, L-tryptophan, p-coumaric acid, ferulic acid and traces of un­ identified flavonoids were detected. Introduction acid in their cell walls (Jackman and Smith, 1996). Phenolic compounds are ubiquitous in the plant Except ferulic acid, also other phenolic acids and kingdom and they have been reported to possess phenolic acid conjugates have been reported in many biological effects. Recently plant phenolics different beetroot materials (Winter and Herr­ have received interest due to their supposed prop­ mann, 1986; Bokern et al ., 1991; Waldron et al., erties in promoting human health and the possi­ 1997; Harborne et al., 1999). The antioxidant activ­ bility to use them as natural food additives, since ity of beet extracts, especially beet peel extracts, phenolic compounds influence the quality, accept­ reported recently increase the interest in beetroot ability and stability of foods by acting as flav- compounds (Vinson et al., 1998; Kähkönen et al., orants, colorants and antioxidants (Decker, 1995). 1999). The red beetroot (Beta vulgaris ) has been culti­ In this paper the effect of extraction method and vated for many hundreds of years in all temperate solvent on the content of red beetroot peel extract climates. Beetroot is used as a vegetable, and its are described. The characterisation of chemical juice and extracts also as traditional medicine, compounds in detail is given for beetroot peel ex­ food colorant and additive to cosmetics (Henry, tract (80% aqueous methanol) by a high-perfor­ 1996; Stuppner and Egger, 1996). The use of beet­ mance liquid chromatography coupled with UV- root as source of colour has focused the beetroot detector, DAD-detector and a mass spectrometer investigations on betalains (red-violet betacyanins equipped with electrospray ion source. and yellow betaxanthins), which are a class of N- containing water-soluble plant pigments. Betalains Materials and Methods are characteristic for the plant order Caryophyl- Chemicals lales (Centrospermae), and there is a mutual ex­ clusion of betalains and anthocyanins. However, Folin-Ciocalteu’s reagent was purchased from other flavonoids coexist with the betalains (Steg- Fluka (Biochemica, Buchs, Switzerland), sodium lich and Strack, 1990). Many betalain-bearing spe­ carbonate and L-tryptophan were from Merck cies have relatively large concentrations of ferulic (Darmstadt, Germany) and gallic acid, ferulic acid 0939-5075/2001/0500-0343 $ 06.00 © 2001 Verlag der Zeitschrift für Naturforschung, Tübingen ■ www.znaturforsch.com ■ D 344 T. Kujala et al. • Red Beetroot Peel Extracts and p-coumaric acid were from Sigma (St. Louis, adjusted to 2.0 using HC1. The preparative column MO, USA). All organic solvents used were of (400 x 25 mm I.D., Pharmacia, Uppsala, Sweden) HPLC-grade. filled with Sephadex LH-20 (Pharmacia, Umeä, Sweden) gel was wrapped in aluminum foil to pre­ Plant material vent exposure to light during the fractionation. The column was washed in advance with water The red beetroots were obtained from a local (pH 2.0) and the extract was transferred on to the farmer (Beta vulgaris , cultivar “Little Ball”). The column. The beetroot peel compounds were beetroots were washed and hand-peeled, and the eluted with acidic water (pH 2.0), neutral water collected peels were cut into small pieces and and different strengths (20-100%) of aqueous stored at -25° C until lyophilisation. Lyophilised methanol. The first fraction contained sugars and plant material was ground with a mortar and pes­ was discarded. The elution was continued and tle. fractions (about 100 ml per fraction) were col­ lected. The fractions were taken almost to dryness Extraction (analytical) by a rotary evaporator and the residues were di­ Three replicates were prepared with both ex­ luted to water in prior to HPLC and HPLC-ESI- traction methods and solvents. MS analyses. Portion of the fractions were lyo­ M ethod A: 500 mg of lyophilised material was philised for further analyses. extracted with 25 ml of solvent (80% aqueous methanol or water) in planar shaker (Promax Determination of total phenolics 2020, Heidolp, Germany) for 45 min. The homoge- The content of total phenolics in the extracts nate was centrifuged for 10 min (1500 xg) and the was determined according to a modification of the clear supernatant was collected. The insoluble part Folin-Ciocalteu method (Nurmi et al., 1996) on a was re-extracted twice with 25 ml of solvent for 45 Perkin-Elmer Lambda 12 UV/VIS Spectrometer and 20 min. The combined extracts were taken to (Norwalk, CT, USA). The total phenolic content dryness by a rotary evaporator and the residue was expressed as gallic acid equivalents (GAE) in was dissolved in 20 ml of water. milligrams per gram of dry plant material. M ethod B\ 500 mg of lyophilised material was homogenised (Ultra-Turrax T 25, Janke & Kunkel, HPLC analysis IKA-Labortechnik, Germany) for one min with 10 ml of solvent (80% aqueous methanol or Analytical chromatographic measurements water). The homogenate was centrifuged for were performed on an HPLC system consisting of 10 min (1500xg) and the clear supernatant was a Merck-Hitachi L-7200 autosampler, Merck-Hi- collected. The insoluble part was re-extracted with tachi L-7100 HPLC pump, Merck-Hitachi L-7400 10 ml of solvent. The combined extracts were UV/VIS detector and Merck-Hitachi D-7000 taken to dryness by a rotary evaporator and the interface (all from Hitachi, Tokyo, Japan). The ab­ residue was dissolved in 20 ml of water. sorption spectra (240-370 nm) of individual com­ pounds of extracts and fractions were recorded on a system consisting the pump and LC-235 diode Extraction (preparative) and fractionation array detector (Perkin-Elmer, Norwalk, CT, USA) Prior to the HPLC-ESI-MS analysis a concen­ connected to a Graphic Printer GP-100 (Perkin- trated beetroot peel extract was prepared. The ex­ Elmer, Beaconsfield, UK). The compounds were tract was prepared by a modification of the analyt­ detected by an HPLC method developed earlier ical method B, the initial amount of dry beetroot in our laboratory (Ossipov et al., 1995). The col­ peel being 20 g and the volume of used solvent, umn used was a LiChrocart RP-18 (250 x 4.0 mm 80% aqueous methanol, 800 ml. The combined ex­ I.D., Purospher, 5-|.im, Merck, Darmstadt, Ger­ tracts were taken to dryness and the residue was many) equipped with a precolumn. The injection dissolved in 40 ml of water. volume was 20 |.il. A constant flow rate was 1.0 ml The concentrated extract was centrifuged for min-1. The eluents were acetonitrile (A) and for­ 10 min (1500xg) and the pH of the extract was mic acid - water (5:95, v/v) (B). The elution pro­ T. Kujala et al. • Red Beetroot Peel Extracts 345 file was: 0-5 min, 100% B (isocratic); 5-60 min, The mass spectrometer was operated in the 0-30% A in B (linear); 60-70 min, 30-60% A in negative and positive ion modes. Mass data were B (linear). In order to define the composition of acquired by the scan mode, consisted of scanning beetroot peel extract, both betalains and pheno­ from m /z 100 to 1100 with 0.30 amu steps. The lics, 280 nm was chosen as detection wavelength, spray needle potential for the negative ion experi­ after also wavelenghts 320, 365 and 520 nm were ments was -4000 V. Orifice plate voltage was set tried. at - 35 V. Ring voltage was set at -220 V. Nebu­ lizer gas (purified air) value was set to value 8 and curtain gas (N2) to value 10. Settings for the posi­ HPLC-ESI-MS analysis tive ion experiments were: the spray needle poten­ A high-performance liquid chromatographic- tial, 5200 V; orifice plate voltage, 45 V; ring volt­ mass spectrometric analysis of red beetroot peel age, 220 V; nebulizer gas, value 8 and curtain gas, compounds were carried out with a Perkin-Elmer value 10. Heated nitrogen gas temperature was SCIEX API 365 triple-quadrupole mass spectrom­ 300 °C. eter (Sciex, Toronto, Canada) incorporated to a Perkin-Elmer Series 200 HPLC system and Apple Results and Discussion Macintosh 8.1 data system. The ion source was an ionspray (pneumatically assisted electrospray ioni­ sation option; Sciex, Toronto, Canada). The HPLC The total phenolic contents of the red beetroot system was comprised of two Perkin-Elmer Series peel extracts prepared by different extraction 200 micro pumps (Perkin-Elmer, Norwalk, CT, methods and solvents decreased in the order 80% USA) and a 785A UV/VIS detector (Perkin-El­ aqueous methanol extract prepared by method A mer, Norwalk, CT, USA).